CN109861838A - The method and system of services topology and service monitoring are generated based on streaming computing - Google Patents
The method and system of services topology and service monitoring are generated based on streaming computing Download PDFInfo
- Publication number
- CN109861838A CN109861838A CN201711238278.2A CN201711238278A CN109861838A CN 109861838 A CN109861838 A CN 109861838A CN 201711238278 A CN201711238278 A CN 201711238278A CN 109861838 A CN109861838 A CN 109861838A
- Authority
- CN
- China
- Prior art keywords
- service
- minute
- index value
- time
- logically
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Abstract
The invention discloses a kind of method and system that services topology and service monitoring are generated based on streaming computing, methods are as follows: receives source traffic, the incidence relation of all service levels is generated according to source traffic；The logically and physically title and incidence relation of all service levels are received, and stores and arrives RDBMS, generates the logically and physically service data of all service levels；The logically and physically service data for receiving all service levels, calculates the modulus end value of year, day, these three time index values of minute and minute index value, monitor control index value is calculated in corresponding minute slot.This programme quasi real time calculates monitor control index value in time slot by streaming computing, establishes the services topology relationship at service-oriented and technology visual angle by defining the corresponding relationship of logical/physical service and Service Instance, enables operation maintenance personnel from the operating status of business and technology visual angle global monitoring complexity telecommunication system.
Description
Technical field
The present invention relates to telecommunications service monitoring fields, especially relate to a kind of based on streaming computing generation services topology
With the method and system of service monitoring.
Background technique
Due to the portfolio of telecommunication system is big, more than application service and application service mostly uses greatly Deployment System For Clusters, in the past
Mostly using each mode for servicing and independently monitoring, it is difficult to quasi real time monitor whole system operation shape from business and the technology visual angle overall situation
State.
Summary of the invention
In order to solve the defect of the above-mentioned prior art, the object of the present invention is to provide one kind to generate service based on streaming computing
The method and system of topology and service monitoring.
In order to achieve the above objectives, the technical scheme is that
A method of services topology and service monitoring are generated based on streaming computing, comprising the following steps:
Source traffic is received, the incidence relation of all service levels is generated according to source traffic；
The logically and physically title and incidence relation of all service levels are received, and stores and arrives RDBMS, generates institute
There is the logically and physically service data of service level；
The logically and physically service data for receiving all service levels, calculates year, day, these three time index values of minute
And the modulus end value of minute index value, monitor control index value is calculated in corresponding minute slot.
Further, the reception source traffic generates the incidence relation step of all service levels according to source traffic,
Including,
Span data item is extracted from source traffic；
It whether identifies in Span data item containing all logical service names and physical services title for servicing levels；
If it does, then generating the incidence relation of all service levels；
If not containing, logical service names and physical services title configured with missing service level, and generate institute
There is the incidence relation of service level.
Further, the logically and physically title for receiving all service levels and incidence relation, and store
To RDBMS, the logically and physically service data step of all service levels is generated, including,
The renewal time of logically and physically service data is set, and judges whether renewal time reaches；
If renewal time reaches, the logically and physically title for servicing level and incidence relation are stored into RDBMS.
Further, it is described receive it is all service levels logically and physically service datas, calculate year, day, minute this
The modulus end value of three time index values and minute index value calculates monitor control index value step in corresponding minute slot,
Including,
Year, day, these three time indexs of minute are calculated according to the timestamp of the logically and physically service data received
The modulus end value of value and minute index value；
By the time index values after calculating respectively compared with the time index values of three time slots, and judge whether identical；
If the time index values after calculating are equal to the time index values of corresponding any time slot, taken with minute index value
Mould；
If the time index values after calculating are greater than the index value of maximum time slot, by the prison of the minute slot of corresponding serial number
It controls index value and is stored in database, and reinitialize the minute slot of the serial number.
Further, described with minute index value removing step, including,
It identifies modulus end value, and monitor control index value is added up and calculated.
A kind of system that services topology and service monitoring are generated based on streaming computing, comprising:
Service analysis generator generates the association of all service levels according to source traffic for receiving source traffic
Relationship；
Service data storage, for receiving the logically and physically title and incidence relation of all service levels,
And store and arrive RDBMS, generate the logically and physically service data of all service levels；
Monitor control index value calculator and memory, for receiving the logically and physically service data of all service levels, meter
The modulus end value for calculating year, day, these three time index values of minute and minute index value, calculates in corresponding minute slot
Monitor control index value.
Further, the service analysis generator includes extraction module, identification module and generation module,
The extraction module, for extracting Span data item from source traffic；
The identification module, for identification in Span data item whether containing it is all service levels logical service names with
And physical services title；
The generation module generates the incidence relation of all service levels, if judgement is not if identification module judgement contains
Contain, then logical service names and physical services title configured with missing service level, and generates all service levels
Incidence relation.
Further, the service data storage includes timing update module, logically and physically services number for setting
According to renewal time, and judge whether renewal time reaches；If renewal time reaches, the logically and physically name of level will be serviced
Claim and incidence relation storage is into RDBMS.
Further, the monitor control index value calculator and memory include computing module, contrast module, modulus module and
Initialization module,
The computing module, for calculating year according to the timestamp of the logically and physically service data received, day, dividing
The modulus end value of these three time index values of clock and minute index value；
The contrast module, for the time index values after calculating respectively with the time index values of three time slots ratio
Compared with, and judge whether identical；
The modulus module, for time index values after computation be equal to corresponding any time slot time index values it
Afterwards, with minute index value modulus；
The initialization module is greater than after the index value of maximum time slot for time index values after computation,
Monitor control index value in the minute slot of corresponding serial number is stored in database, and reinitializes serial number minute slot.
Further, the modulus module, modulus end value for identification, and monitor control index value is added up and counted
It calculates.
The beneficial effects of the present invention are: scheme quasi real time calculates monitor control index value in time slot, logical by streaming computing
It crosses and defines the corresponding relationship of logical/physical service and Service Instance and establish the services topology relationship at service-oriented and technology visual angle,
Enable operation maintenance personnel from the operating status of business and technology visual angle global monitoring complexity telecommunication system；By the way of time slot
Monitor control index value is calculated, so that the accuracy of monitor control index value is higher, monitor control index value is made to be limited to less memory space
It is interior.
Detailed description of the invention
Fig. 1 is a kind of method flow that services topology and service monitoring are generated based on streaming computing of one embodiment of the invention
Figure；
Fig. 2 is a kind of method flow diagram for the incidence relation for generating all service levels of one embodiment of the invention；
Fig. 3 is the method flow diagram that a kind of timing of one embodiment of the invention updates logically and physically service data；
Fig. 4 is a kind of method flow diagram of monitor control index value calculating method of one embodiment of the invention；
Fig. 5 is a kind of method flow diagram of monitor control index value calculating method of another embodiment of the present invention；
Fig. 6 is the monitor control index value exemplary diagram in a kind of time slot of one embodiment of the invention；
Fig. 7 is a kind of framework for the system that services topology and service monitoring are generated based on streaming computing of one embodiment of the invention
Figure；
Fig. 8 is a kind of logic for the system that services topology and service monitoring are generated based on streaming computing of one embodiment of the invention
Flow chart；
Fig. 9 is a kind of structural block diagram of service analyser of one embodiment of the invention；
Figure 10 is a kind of structural block diagram of service data storage of one embodiment of the invention；
Figure 11 is a kind of structural block diagram of monitor control index value calculator and memory of one embodiment of the invention.
Specific embodiment
To illustrate thought and purpose of the invention, the present invention is done further below in conjunction with the drawings and specific embodiments
Explanation.
RDBMS, full name in English: relational database management system, Chinese: relationship type
Data base management system.RDBMS is stored and is managed by data, relationship and to the data model of the constraint three composition of data
Data.
It calculates in real time, it is emphasised that in real time.For example Xiao Ming will check the consumption full amount of his the whole of last year, then working as Xiao Ming
When point lower statistics button, just the crack in Pi loses no time to calculate server cluster, it is necessary to when Xiao Ming can restrain oneself
Between obtain a result in range.Realize that typically redundancy+various high performance units are supporting, algorithm in the behind of this calculating
Also real-time is optimized, but calculates do not emphasize with that algorithm in real time, as long as can guarantee high real-time.It is real
When calculate and distinguished with the maximum of off-line calculation, be exactly off-line calculation be to calculate the time that people is out of patience, because this person is not required to
It waits, task is lost to computer.
Streaming computing, it is slightly more blunt than calculating in real time but again more in real time than off-line calculation, and mainly emphasize
It is calculation method.For example, server end, there is a value, it is record Xiao Ming's quantity on order.After Xiao Ming often buys a thing, clothes
As soon as business end issues the successful event of transaction immediately, which adds 1 after receiving this event immediately.If with off-line calculation
Mode is done, estimation be in inquiry, it is just sluggish from low speed storage, all orders of Xiao Ming are taken out, statistical number
Amount.Streaming computing is somewhat like the trigger of database field, and something like the calculating mode of event bus, middleware etc.
Referring to figs. 1 to Fig. 6, a specific embodiment of the invention is proposed, one kind generating services topology and clothes based on streaming computing
The method for being engaged in monitoring, comprising the following steps:
S10, source traffic is received, the incidence relation of all service levels is generated according to source traffic.
S20, the logically and physically title and incidence relation for receiving all service levels, and store and arrive RDBMS, it is raw
At the logically and physically service data of all service levels.
S30, the logically and physically service data for receiving all service levels, calculate year, day, these three time ropes of minute
The modulus end value for drawing value and minute index value calculates monitor control index value in corresponding minute slot.
For step S10, after receiving source traffic, physics at different levels and logical services name are obtained from source traffic
Claim, part of physics and logical service names can be obtained directly from source traffic, part physical and logical service names
It is not directly present in source traffic, needs to obtain by certain configuration.
With reference to Fig. 2, step S10 the following steps are included:
S11, Span data item is extracted from source traffic.
Whether contain all logical service names and physical services name for servicing levels in S12, identification Span data item
Claim.
S13, if it does, then generate it is all service levels incidence relations.
If S14, not containing, logical service names and physical services title configured with missing service level, and it is raw
At the incidence relation of all service levels.
For step S11, after receiving source traffic, Span data item is extracted from source traffic, the data item
At least containing TraceId, SpanId, ParentId, HostIp, InstanceName, CreatTime, CallTime,
The fields such as IsEntry, ErrCode, wherein InstanceName is Service Instance title, it is both 2 grades of logical service names,
It is 2 grades of physical services titles, InstantceName is updated into database.
For step S12-S14, identify in Span data item whether the logical service names containing all service levels with
And physical services title, there is the logical service names met and physical services title, call directly, not so by being arranged to
To the logical service names and physical services title of the service level of missing.
For example, being generated if 1 grade of logical service names ComponentName can be extracted from Span data item
The incidence relation data of ComponentName and InstanceName；Otherwise, by configuring to obtain 1 grade of logical service names
ComponentName, and generate the incidence relation data of ComponentName and InstanceName.In addition, in source traffic
Containing HostIp, it has had the incidence relation with InstanceName as 1 grade of physical services title.
Meanwhile if 0 grade of logical service names AppName can be extracted from source traffic, generate AppName with
The incidence relation data of ComponentName.Otherwise, it by configuring to obtain 0 grade of logical service names AppName, and generates
The incidence relation data of AppName and ComponentName are stored into database.Wherein, AppName can be used as 0 grade simultaneously
Physical services title.
Finally obtain 2 grades, 1 grade, 0 grade of logical/physical service name and incidence relation, the pass as all service levels
Join relation data, in case subsequent use.
Specifically, instanceName is used (to make in the present invention as the shared base-level of physical topology, logical topology
To default 2 grades of level) so that physical topology, logical topology establish incidence relation from bottom service；By defining logic
The corresponding relationship of service and one or more physical services, it is established that the services topology relationship of service-oriented；Meanwhile it using
InstanceName (as 2 grades of level of default in the present invention), HostIp (as 1 grade of level of default in the present invention) are used as two-stage
Physical services rank, it is established that the services topology relationship of technology oriented enables operation maintenance personnel global from business and technology visual angle
Monitor the operating status of complicated telecommunication system.
For step S20, the logically and physically title and incidence relation of all service levels are received, and stores and arrives
RDBMS generates the logically and physically service data of all service levels.Such as receive above-mentioned 2 grades, 1 grade, 0 grade of logic/object
Service name and incidence relation are managed, further arrives 2 grades, 1 grade, 0 grade of logical/physical service name and incidence relation storage
In RDBMS, the logically and physically service data of three-level is formed, specifically, logically and physically service data is logical/physical service
Topological diagram data.
Logical services are not limited only to 2,1,0 this three-level, more stages logical topological structure can be also extended according to practical business, only
It need to obtain or configure from source traffic and obtain.For physical services other than 0 grade is to define, 1 grade and 2 grades is all true
It is existing, as long as there is the physical services level of physical presence, so that it may according to true extension is more and physical topological structure, at present only
1 grade and 2 grades is that physics is really existing, and 0 grade is definition, if continuing to define on the basis of 0 grade, be can be
Continue to define the ranks such as -1 grade, -2 grades.Specifically, be then 0 grade most top service definition in order not to negative level occur,
Other ranks are incremented by the ranks such as 1 grade, 2 grades, 3 grades with this.
Referring to Fig. 3, step S20 the following steps are included:
S21, the renewal time for setting logically and physically service data, and judge whether renewal time reaches.
If S22, renewal time reach, logically and physically service data is updated.
For step S21, a specific renewal time, such as 10 minutes, 5 minutes or 3 minutes are set, as more new logic
Logically and physically service data is automatically updated when reaching renewal time with the time interval of physical service data.So can be with
By judging whether renewal time reaches, to determine whether to update logically and physically service data.
Whether renewal time reaches, and can be and is realized by judging whether time point is corresponding renewal time point.Example
Such as, it is divided between renewal time 10 minutes, last renewal time is 15 minutes, just by judging whether current time is 25 points
Clock, 35 minutes, 45 minutes, 55 minutes or 5 minutes are to determine whether more new data then determines more if any of the above-described time point
The new time has reached, and automatically updates logically and physically service data.
For step S22, determines that renewal time reaches in S21 step, then update logically and physically service data, generate new
Service logic and physical service data.
For step S30, after often receiving primary logically and physically service data, logically and physically service data includes
Having time stabs information, calculates year, day, these three time index values of minute and minute index value according to the timestamp of data
Modulus end value calculates monitor control index value in corresponding minute slot.
Time slot is for storing year index value, day index value, minute index value, monitor control index value.Year rope in time slot
Draw value, day index value, minute index value, triple combination constitutes slot mark.
Referring to Fig. 4, step S30 the following steps are included:
S31, year, day, these three times of minute are calculated according to the timestamp of the logically and physically service data received
The modulus end value of index value and minute index value.
S32, by the time index values after calculating respectively compared with the time index values of three time slots, and judge whether phase
Together.
If the time index values after S33, calculating are equal to the time index values of corresponding any time slot, with minute index value
Modulus.
If the time index values after S34, calculating are greater than the index value of maximum time slot, the minute slot of serial number will be corresponded to
Monitor control index value be stored in database, and reinitialize the minute slot of the serial number.
For step S31, time slot is for storing year index value, day index value, minute index value, monitor control index value.When
Between in slot year index value, day index value, minute index value, triple combination constitutes slot mark.The meter by the way of time slot
Monitor control index value is calculated, so that the accuracy of monitor control index value is higher, is limited to monitor control index value in less memory space.
Year index value refers to year value, day index value refer to this day is which day in 1 year, minute index value refers to this
Which the number of minutes be in one day minute.For example on the June 23rd, 2017 of 1:00 AM 18 divides, a year index value is 2017, and day indexes
Value is 174, and minute index value is 78, and the end value of 78 mould 3 of minute slot index value is 0, indicates that the data of this minute are stored in
In 0th minute slot.The corresponding year index value of timestamp temporal information, day index value and minute index are obtained by streaming computing
Value is used for subsequent use.
For step S32, by by the time index values after calculating respectively compared with the time index values of three time slots,
Just can determine that whether the time index values of time index values and time slot after calculating are consistent.Only newly come in logically and physically to take
Be engaged in years of data, day, these three index values corresponding with time slot year minute, day, these three index values of minute it is equal when, just meeting
The monitoring value for the logically and physically service data of newly coming in is added in the existing monitor control index value of minute slot.
For step S33, when step S32 determines that time index values after calculating are equal to the time rope of corresponding any time slot
Draw value, then with minute index value modulus.Time index values after determining calculating are not equal to the time rope of corresponding any time slot
When drawing value, then the monitoring value for the logically and physically service data of newly coming in will not be added in the existing monitoring value of minute slot.
Step S33, comprising steps of S330, identification modulus end value, and monitor control index value is added up and calculated.
For step S330, according to the size of the modulus end value recognized to the monitor control index value in corresponding minute slot into
Row is cumulative and calculates.The number of time slot is the sending cycle of modulus and monitor control index value, in quasi real time scene, can use N
Minute is a sending cycle, that is, is corresponding with N number of time slot.Serial number of the operation result of minute index value %N as time slot,
Each data need to calculate monitoring value in corresponding time slot by minute index value.Wherein, N is natural number, can be as needed
With other several modulus, such as 1,2,3,4.
It, then will corresponding sequence when time index values for step S34, after calculating are greater than the index value of maximum time slot
The monitor control index value of the minute slot of number (the modulus end value that serial number is minute index value) is stored in database, and reinitializes this
Serial number minute slot.The original monitor control index value being present in minute slot is removed.
Be the monitor control index value example in time slot with reference to Fig. 6, specifically, monitor control index value include TPS, error number,
Average response time.
TPS, Chinese are handling capacity, often receive a service data just from increasing 1.
Error number, English are Err, often receive an exception service data just from increasing 1.
Average response time, English is CallTimeAvg, as soon as it is subtracted after the response time for often receiving a service data
Average response time in minute slot, obtained difference are again the TPS comprising current this service data divided by TPS, TPS, i.e.,
Increase 1 certainly after receiving the service data, for obtained quotient along with the former average response time in minute slot, end value is exactly flat
The equal response time.Calculation formula is as follows:
CallTimeAvgNew=(CallTime-CallTimeAvg)/TPS+CallTimeAvg
Such as: existing handling capacity TPS is 2 in time slot, and average response time CallTimeAvg is 2.5 milliseconds, now receives 1
New data, response time CallTime is 7 milliseconds, then TPS is 3 from after increasing 1, and new average response time is
/ 3+2.5 milliseconds of CallTimeAvgNew=(7 milliseconds -2.5 milliseconds) is as a result equal to 4 milliseconds, i.e., new average response time value
It is 4 milliseconds.
In Fig. 6, the TPS of leftmost side time slot is 10, error number 3, average response time 95；Interlude slot
TPS is 8, error number 1, average response time 90；The TPS of rightmost side time slot is 6, error number 2, average response time
It is 100.
With reference to Fig. 5, in a specific embodiment of the invention, 3 minutes are taken as a sending cycle, i.e., there are three time slots.
Serial number of the operation result of minute index value %3 as time slot, each data were needed by minute index value in the corresponding time
Monitor control index value is calculated in slot.Specific step is as follows:
S31, year, day, these three times of minute are calculated according to the timestamp of the logically and physically service data received
The modulus end value of index value and minute index value.
S32, by the time index values after calculating respectively compared with the time index values of three time slots, and judge whether phase
Together.
If the time index values after S33, calculating are equal to the time index values of corresponding any time slot, with minute index value
Modulus.
If S331, result are equal to 0, the monitor control index value of minute slot 0 is added up and is calculated.
If S332, result are equal to 1, the monitor control index value of minute slot 1 is added up and is calculated.
If S333, result are equal to 2, the monitor control index value of minute slot 2 is added up and is calculated.
If the time index values after S341, calculating are greater than the index value of maximum time slot, the minute of serial number will be corresponded to
The monitor control index value of slot (slot serial number 0 or 1 or 2) is put in storage and reinitializes serial number minute slot (slot serial number 0 or 1 or 2).
When the data of second period reach, minute index value modulus end value can be any one time slot sequence
Number, the monitor control index value of corresponding time slot serial number sending time slot empties time slot data and saves the monitor control index of new data
Value.Wherein, serial number is the modulus end value of minute index value.
If more than two periods still without new data reach, in order to avoid monitor control index value update not in time, then
Time slot monitoring data more than two periods sends and empties time slot data.
This programme quasi real time calculates monitor control index value in time slot by streaming computing, by defining logical/physical clothes
The corresponding relationship of business and Service Instance establishes the services topology relationship at service-oriented and technology visual angle, and operation maintenance personnel is obtained employment
The operating status of business and technology visual angle global monitoring complexity telecommunication system；Monitor control index value is calculated by the way of time slot, is made
Monitor control index value accuracy it is higher, monitor control index value is limited in less memory space.
With reference to Fig. 7-Figure 11, it is based on what streaming computing generated services topology and service monitoring the invention also provides a kind of
System, comprising:
Service analysis generator 10 generates the pass of all service levels according to source traffic for receiving source traffic
Connection relationship.
Service data storage 20, logically and physically title and association for receiving all service levels are closed
System, and store and arrive RDBMS, generate the logically and physically service data of all service levels.
Monitor control index value calculator and memory 30, for receiving the logically and physically service data of all service levels,
The modulus end value for calculating year, day, these three time index values of minute and minute index value is fallen into a trap in corresponding minute slot
Calculate monitor control index value.
Referring to Fig. 7, source traffic belongs to source data fluid layer, service analysis generator 10, service data storage 20, monitoring
Value calculator and memory 30 are located at streaming computing layer, and RDBMS and time series database are located at data storage layer.Referring to Fig. 8, source number
According to stream be sent to after service analysis generator 10 by service packet be sent to service data storage 20 and monitoring value calculator with
Services topology is finally stored in RDBMS by memory 30, and monitor control index value is stored in time series database.
For service analysis generator 10, after receiving source traffic, Span data item is extracted from source traffic,
The data item at least contain TraceId, SpanId, ParentId, HostIp, InstanceName, CreatTime,
The fields such as CallTime, IsEntry, ErrCode, wherein InstanceName is Service Instance title, it is both 2 grades of logic clothes
Title of being engaged in and 2 grades of physical services titles update InstantceName into database.
With reference to Fig. 9, service analysis generator 10 includes extraction module 11, identification module 12 and generation module 13.
Extraction module 11, for extracting Span data item from source traffic.
Identification module 12, for identification in Span data item whether containing it is all service levels logical service names and
Physical services title.
Generation module 13 generates the incidence relation of all service levels, if judgement is not if the judgement of identification module 12 contains
Contain, then logical service names and physical services title configured with missing service level, and generates all service levels
Incidence relation.
Cooperated by extraction module 11, identification module 12 and generation module 13 and realizes whether contain in identification Span data item
, there are the logical service names met and physical services in the logical service names and physical services title of all service levels
Title calls directly, and not so passes through the logical service names and physical services title for configuring the service level lacked.
For example, being generated if 1 grade of logical service names ComponentName can be extracted from Span data item
The incidence relation data of ComponentName and InstanceName；Otherwise, by configuring to obtain 1 grade of logical service names
ComponentName, and generate the incidence relation data of ComponentName and InstanceName.In addition, in source traffic
Containing HostIp, it has had the incidence relation with InstanceName as 1 grade of physical services title.
Meanwhile if 0 grade of logical service names AppName can be extracted from source traffic, generate AppName with
The incidence relation data of ComponentName.Otherwise, it by configuring to obtain 0 grade of logical service names AppName, and generates
The incidence relation data of AppName and ComponentName are stored into database.Wherein, AppName can be used as 0 grade simultaneously
Physical services title.
Finally obtain 2 grades, 1 grade, 0 grade of logical/physical service name and incidence relation, the pass as all service levels
Join relation data, in case subsequent use.
For service data storage 20, the logically and physically title and association for receiving all service levels are closed
System, and store and arrive RDBMS, generate the logically and physically service data of all service levels.Such as receive above-mentioned 2 grades, 1
Grade, 0 grade of logical/physical service name and incidence relation, further by 2 grades, 1 grade, 0 grade of logical/physical service name and association
Relationship is stored into RDBMS, forms the logically and physically service data of three-level, specifically, logically and physically service data is to patrol
Volume/physical services topology diagram data.
Logical services are not limited only to 2,1,0 this three-level, more stages logical topological structure can be also extended according to practical business, only
It need to obtain or configure from source traffic and obtain.For physical services other than 0 grade is to define, 1 grade and 2 grades is all true
It is existing, as long as there is the physical services level of physical presence, so that it may according to true extension is more and physical topological structure, at present only
1 grade and 2 grades is that physics is really existing, and 0 grade is definition, if continuing to define on the basis of 0 grade, be can be
Continue to define the ranks such as -1 grade, -2 grades.Specifically, be then 0 grade most top service definition in order not to negative level occur,
Other ranks are incremented by the ranks such as 1 grade, 2 grades, 3 grades with this.
With reference to Figure 10, service data storage 20 includes timing update module 21, logically and physically services number for setting
According to renewal time, and judge whether renewal time reaches；If renewal time reaches, the logically and physically name of level will be serviced
Claim and incidence relation storage is into RDBMS.
For timing update module 21, a specific renewal time, such as 10 minutes, 5 minutes or 3 minutes are set, as
The time interval for updating logically and physically service data automatically updates logically and physically service data when reaching renewal time.
So can be by judging whether renewal time reaches, to determine whether to update logically and physically service data.
Whether renewal time reaches, and can be and is realized by judging whether time point is corresponding renewal time point.Example
Such as, it is divided between renewal time 10 minutes, last renewal time is 15 minutes, just by judging whether current time is 25 points
Clock, 35 minutes, 45 minutes, 55 minutes or 5 minutes are to determine whether more new data then determines more if any of the above-described time point
The new time has reached, and automatically updates logically and physically service data.
Determine that renewal time reaches, then updates logically and physically service data, generate new service logic and physical services
Data.
Monitor control index value calculator and memory 30 are patrolled after often receiving primary logically and physically service data
It collects and physical service data includes timestamp information, year, day, these three time ropes of minute are calculated according to the timestamp of data
The modulus end value for drawing value and minute index value calculates monitor control index value in corresponding minute slot.
Time slot is for storing year index value, day index value, minute index value, monitor control index value.Year rope in time slot
Draw value, day index value, minute index value, triple combination constitutes slot mark.
With reference to Figure 11, monitor control index value calculator and memory 30 include computing module 31, contrast module 32, modulus module
33 and initialization module 34,
Computing module 31, for calculating year, day, minute according to the timestamp of the logically and physically service data received
The modulus end value of these three time index values and minute index value；
Contrast module 32, for by the time index values after calculating respectively compared with the time index values of three time slots,
And judge whether identical；
Modulus module 33, for time index values after computation be equal to corresponding any time slot time index values it
Afterwards, with minute index value modulus；
Initialization module 34 is greater than after the index value of maximum time slot for time index values after computation, will
Monitor control index value in the minute slot of corresponding serial number (the modulus end value that serial number is minute index value) is stored in database, and again
Initialize serial number minute slot.
For computing module 31, time slot is for storing year index value, day index value, minute index value, monitor control index value.
In time slot year index value, day index value, minute index value, triple combination constitute slot mark.
Year index value refers to year value, day index value refer to this day is which day in 1 year, minute index value refers to this
Which the number of minutes be in one day minute.For example on the June 23rd, 2017 of 1:00 AM 18 divides, a year index value is 2017, and day indexes
Value is 174, and minute index value is 78, and the end value of 78 mould 3 of minute slot index value is 0, indicates that the data of this minute are stored in
In 0th minute slot.The corresponding year index value of timestamp temporal information, day index value and minute index are obtained by streaming computing
Value is used for subsequent use.
For contrast module 32, by the time index values after calculating respectively with the time index values of three time slots ratio
Compared with whether the time index values of time index values and time slot after just can determine that calculating are consistent.Logic of only newly coming in and object
Manage year of service data, day, these three index values corresponding with time slot year minute, day, these three index values of minute it is equal when,
The monitoring value for the logically and physically service data of newly coming in can be just added in the existing monitor control index value of minute slot.
For modulus module 33, when contrast module 32 determines that time index values after calculating are equal to corresponding any time slot
Time index values, then with minute index value modulus.Any time slot is corresponded to when determining that time index values after calculating are not equal to
When time index values, then will not be added to the monitoring value for the logically and physically service data of newly coming in the existing monitoring value of minute slot
In.
Specifically, modulus module 33, modulus end value for identification, and monitor control index value is added up and calculated.
The monitor control index value in corresponding minute slot is added up and calculated according to the size of the modulus end value recognized.
The number of time slot is the sending cycle of modulus and monitor control index value, in quasi real time scene, can use N minutes as a hair
The period is sent, that is, is corresponding with N number of time slot.Serial number of the operation result of minute index value %N as time slot, each data need
Monitoring value is calculated in corresponding time slot by minute index value.Wherein, N is natural number, can as needed with other several modulus,
Such as 1,2,3,4.
For initialization module 34, when the time index values after calculating are greater than the index value of maximum time slot, then will
The monitor control index value of the minute slot of corresponding serial number (the modulus end value that serial number is minute index value) is stored in database, and again just
The beginningization serial number minute slot.The original monitor control index value being present in minute slot is removed.
This programme quasi real time calculates monitor control index value in time slot by streaming computing, by defining logical/physical clothes
The corresponding relationship of business and Service Instance establishes the services topology relationship at service-oriented and technology visual angle, and operation maintenance personnel is obtained employment
The operating status of business and technology visual angle global monitoring complexity telecommunication system；Monitor control index value is calculated by the way of time slot, is made
Monitor control index value accuracy it is higher, monitor control index value is limited in less memory space.
The above description is only a preferred embodiment of the present invention, is not intended to limit the scope of the invention, all utilizations
Equivalent structure or equivalent flow shift made by description of the invention and accompanying drawing content is applied directly or indirectly in other correlations
Technical field, be included within the scope of the present invention.
Claims (10)
1. a kind of method for generating services topology and service monitoring based on streaming computing, which comprises the following steps:
Source traffic is received, the incidence relation of all service levels is generated according to source traffic；
The logically and physically title and incidence relation of all service levels are received, and stores and arrives RDBMS, generates all clothes
The logically and physically service data for level of being engaged in；
Receive it is all service levels logically and physically service datas, calculate year, day, these three time index values of minute and
The modulus end value of minute index value, calculates monitor control index value in corresponding minute slot.
2. the method for generating services topology and service monitoring based on streaming computing as described in claim 1, which is characterized in that institute
Reception source traffic is stated, the incidence relation step of all service levels is generated according to source traffic, including,
Span data item is extracted from source traffic；
It whether identifies in Span data item containing all logical service names and physical services title for servicing levels；
If it does, then generating the incidence relation of all service levels；
If not containing, logical service names and physical services title configured with missing service level, and generate all clothes
The incidence relation for level of being engaged in.
3. the method for generating services topology and service monitoring based on streaming computing as described in claim 1, which is characterized in that institute
The logically and physically title and incidence relation for receiving all service levels are stated, and stores and arrives RDBMS, generates all services
The logically and physically service data step of level, including,
The renewal time of logically and physically service data is set, and judges whether renewal time reaches；
If renewal time reaches, the logically and physically title for servicing level and incidence relation are stored into RDBMS.
4. the method for generating services topology and service monitoring based on streaming computing as described in claim 1, which is characterized in that institute
The logically and physically service data for receiving all service levels is stated, year, day, these three time index values of minute are calculated and is divided
The modulus end value of clock index value calculates monitor control index value step in corresponding minute slot, including,
According to the timestamp of the logically and physically service data received calculate year, day, minute these three time index values with
And the modulus end value of minute index value；
By the time index values after calculating respectively compared with the time index values of three time slots, and judge whether identical；
If the time index values after calculating are equal to the time index values of corresponding any time slot, with minute index value modulus；
If the time index values after calculating are greater than the index value of maximum time slot, the monitoring of the minute slot of corresponding serial number is referred to
Scale value is stored in database, and reinitializes serial number minute slot.
5. the method for generating services topology and service monitoring based on streaming computing as described in claim 1, which is characterized in that institute
It states with minute index value removing step, including,
It identifies modulus end value, and monitor control index value is added up and calculated.
6. a kind of system for generating services topology and service monitoring based on streaming computing characterized by comprising
Service analysis generator generates the incidence relation of all service levels according to source traffic for receiving source traffic；
Service data storage for receiving the logically and physically title and incidence relation of all service levels, and is deposited
RDBMS is stored up, the logically and physically service data of all service levels is generated；
Monitor control index value calculator and memory are calculated for receiving the logically and physically service data of all service levels
Year, day, these three time index values of minute and minute index value modulus end value, monitoring is calculated in corresponding minute slot
Index value.
7. the system for generating services topology and service monitoring based on streaming computing as claimed in claim 6, which is characterized in that institute
Stating service analysis generator includes extraction module, identification module and generation module,
The extraction module, for extracting Span data item from source traffic；
Whether the identification module contains all logical service names and object for servicing levels in Span data item for identification
Manage service name；
The generation module generates the incidence relation of all service levels, if judgement is free of if identification module judgement contains
Have, then logical service names and physical services title configured with missing service level, and generates the pass of all service levels
Connection relationship.
8. the system for generating services topology and service monitoring based on streaming computing as claimed in claim 6, which is characterized in that institute
Stating service data storage includes timing update module, for setting the renewal time of logically and physically service data, and is judged
Whether renewal time reaches；If renewal time reaches, the logically and physically title for servicing level and incidence relation are stored
Into RDBMS.
9. the system for generating services topology and service monitoring based on streaming computing as claimed in claim 6, which is characterized in that institute
It states monitor control index value calculator and memory includes computing module, contrast module, modulus module and initialization module,
The computing module, for according to the timestamp of logically and physically service data received calculate year, day, minute this
The modulus end value of three time index values and minute index value；
The contrast module, for by the time index values after calculating respectively compared with the time index values of three time slots, and
Judge whether identical；
The modulus module is equal to after the time index values of corresponding any time slot for time index values after computation,
With minute index value modulus；
The initialization module is greater than after the index value of maximum time slot for time index values after computation, will be right
It answers the monitor control index value in the minute slot of serial number to be stored in database, and reinitializes serial number minute slot.
10. the system for generating services topology and service monitoring based on streaming computing as claimed in claim 6, which is characterized in that
The modulus module, modulus end value for identification, and monitor control index value is added up and calculated.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711238278.2A CN109861838B (en) | 2017-11-30 | 2017-11-30 | Method and system for generating service topology and monitoring service based on stream computing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711238278.2A CN109861838B (en) | 2017-11-30 | 2017-11-30 | Method and system for generating service topology and monitoring service based on stream computing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109861838A true CN109861838A (en) | 2019-06-07 |
| CN109861838B CN109861838B (en) | 2022-02-22 |
Family
ID=66888101
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711238278.2A Active CN109861838B (en) | 2017-11-30 | 2017-11-30 | Method and system for generating service topology and monitoring service based on stream computing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109861838B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101552739A (en) * | 2008-04-03 | 2009-10-07 | 北京邮电大学 | Method and system for managing life cycle of network service process example |
| CN102143220A (en) * | 2010-02-02 | 2011-08-03 | 国际商业机器公司 | Method and system for discovering physical server location by correlating external and internal server information |
| US20150071109A1 (en) * | 2013-09-09 | 2015-03-12 | Grigori Dzekon | Methods and Systems for Packet Delivery Based Upon Embedded Instructions |
| WO2016140477A1 (en) * | 2015-03-01 | 2016-09-09 | 엘지전자 주식회사 | Broadcasting signal transmission apparatus, broadcasting signal reception apparatus, broadcasting signal transmission method, and broadcasting signal reception method |
| CN107305518A (en) * | 2016-04-20 | 2017-10-31 | 中兴通讯股份有限公司 | The display methods and device of warning information |
-
2017
- 2017-11-30 CN CN201711238278.2A patent/CN109861838B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101552739A (en) * | 2008-04-03 | 2009-10-07 | 北京邮电大学 | Method and system for managing life cycle of network service process example |
| CN102143220A (en) * | 2010-02-02 | 2011-08-03 | 国际商业机器公司 | Method and system for discovering physical server location by correlating external and internal server information |
| US20150071109A1 (en) * | 2013-09-09 | 2015-03-12 | Grigori Dzekon | Methods and Systems for Packet Delivery Based Upon Embedded Instructions |
| WO2016140477A1 (en) * | 2015-03-01 | 2016-09-09 | 엘지전자 주식회사 | Broadcasting signal transmission apparatus, broadcasting signal reception apparatus, broadcasting signal transmission method, and broadcasting signal reception method |
| CN107305518A (en) * | 2016-04-20 | 2017-10-31 | 中兴通讯股份有限公司 | The display methods and device of warning information |
Non-Patent Citations (1)
| Title |
|---|
| 李俊炜: "基于Oracle数据库抵效语句监控方法的研究与应用", 《中国优秀硕士学位论文全文数据库（电子期刊）》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109861838B (en) | 2022-02-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110865929B (en) | Abnormality detection early warning method and system | |
| Karagiannis et al. | A nonstationary Poisson view of Internet traffic | |
| CN107608862B (en) | Monitoring alarm method, monitoring alarm device and computer readable storage medium | |
| CN106126391A (en) | System monitoring method and apparatus | |
| CN108494810A (en) | Network security situation prediction method, apparatus and system towards attack | |
| CN105071983A (en) | Abnormal load detection method for cloud calculation on-line business | |
| CN104917627B (en) | A kind of log cluster for large server cluster scans and analysis method | |
| CN109376924A (en) | A kind of method, apparatus, equipment and the readable storage medium storing program for executing of material requirements prediction | |
| CN106909117B (en) | Data real-time monitoring system and method | |
| CN111352808A (en) | Alarm data processing method, device, equipment and storage medium | |
| WO2018236364A1 (en) | Generating wireless network access point models using clustering techniques | |
| CN109739919A (en) | A kind of front end processor and acquisition system for electric system | |
| CN109901927A (en) | Intelligent Task dynamic scheduling system and its method | |
| US20170262773A1 (en) | Method and system for generating a training model for fabricating synthetic data and for fabricating synthetic data | |
| CN107018408A (en) | The Quality of experience appraisal procedure of mobile terminal HTTP video flowings | |
| CN107291594A (en) | The device and method that openstack platforms are monitored and managed to ceph | |
| CN110297742A (en) | Data monitoring system, method and server | |
| Yamada et al. | Feature-selection based data prioritization in mobile traffic prediction using machine learning | |
| CN104572298B (en) | The resource regulating method and device of video cloud platform | |
| CN105357637B (en) | A kind of position and behavioural information forecasting system and method | |
| CN113839835A (en) | Top-k flow accurate monitoring framework based on small flow filtering | |
| CN108376292A (en) | A kind of crowd's method for predicting, system and equipment | |
| CN106502856B (en) | A kind of information processing method and device | |
| CN114721898A (en) | Edge cloud server utilization rate prediction method and device based on boosting algorithm and storage medium | |
| CN109861838A (en) | The method and system of services topology and service monitoring are generated based on streaming computing |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |